"Misunderstanding of the lambda calculus"

[Chong Yidong]

This may amuse some of the readers of this mailing list.  While
browsing the Wikipedia entry on lambda calculus, I saw that some wit
had written:

  Lisp uses a variant of lambda notation for defining functions, but
  only its purely functional subset is really equivalent to lambda
  calculus. Strictly speaking, this holds only for modern dialects of
  Lisp, such as Common Lisp and Scheme. More archaic Lisps, such as
  Emacs Lisp, still use dynamic binding, and so are not based on the
  lambda calculus. Rather, they are based on the syntax of the lambda
  calculus, together with a misunderstanding of the notion of binding
  and substitution in the lambda calculus.